A subset of tumor cells have the unique ability to self-renew and are responsible for relapse. Uncovering the cell types and mechanisms that govern self-renewal will be imperative for designing new therapies in the treatment of cancer. The focus of this application is to uncover the mechanisms involved in cancer cell self-renewal in embryonal rhabdomyosarcoma (ERMS) - a devastating pediatric cancer of muscle. Having developed a robust model of zebrafish ERMS that accurately recapitulates the molecular underpinnings of this disease and identified the myf5-expressing cancer-initiating cell, experiments outlined here will uncover the basic biology of self-renewal and determine if MYF5 is required for tumor growth and self-renewal. We hypothesize that ERMS results from acquisition of self- renewal potential that is associated with the MYF5 gene program. The goal of aim 1 will be to visualize self-renewal in embryonal rhabdomyosarcoma. Using confocal imaging and transgenic reporter lines that express fluorescent proteins in various stages of muscle development, we now have unprecedented and unfettered access to visualize tumor-initiating cells in their endogenous niche. Such experiments have never been described in any model of cancer. Experiments outlined in aim 1 will determine if self-renewal is confined to specific tumor niches associated with tumor growth, the extent to which asymmetric and symmetric cell division occurs in self-renewing cancer cells at various stables of tumorigenesis, and if ERMS cells can de-differentiate from more mature cell populations into self-renewing myf5+ tumor-initiating cells. These experiments capitalize on the unique attributes of the zebrafish model system to directly visualize cancer processes in vivo by time- lapse confocal and 2-photon imaging. Aim 2 will determine if MYF5 regulates self-renewal potential in both zebrafish and human ERMS. Microarray expression profiling shows that MYF5 is highly expressed in human ERMS when compared to alveolar RMS and normal muscle, suggesting that MYF5+ cells are abundant in this subtype of disease. Using the zebrafish ERMS cancer model in conjunction with human cell culture experiments, we will assess if MYF5 is required for tumor growth and self-renewal. Similarly, we will also assess if MYF5 is an oncogene and can lock cells in a self-renewing cell state that predisposes to ERMS. In total, this application provides a comprehensive plan to interrogate self-renewal in ERMS with the long-term goal of this work to identify molecular pathways associated with self- renewal and aggression, ultimately leading to the identification of drug targets for the treatment of ERMS.